McGraw-Hill

316 ExpandingOpportunities
in 1987, when the market fell by about 20 percent. Assuming the
longs and shorts move in line, the value of the long positions will
decline from $90 to $72, for a loss of $18, and the value of the short
positions will also decline from $90 to $72 (but for gain a of $18). The
securities' lenders are now overcollateralized and will transfer $18
to the long-short account, increasing the liquidity buffer to $28. A
crash, in effect, creates liquidity for a long-short portfolio!
Myth 11. Long-short portfolios are infinitely riskier than long-only
portfolios because losses on short positions are unlimited.
Whereas the risk to a long investment in a security is limited
because the price of the security can go to zero but not below, the
risk of a short position is theoretically unlimited because is there no
bound on rise a in the security's price. The risk of a precipitous rise,
or gap-up, in a security's price is a consideration, but is it one that is
tempered in the context of a portfolio diversified across many securities.
The prices of all the securities sold short are unlikely to rise
dramatically at the same time with no offsetting increases in the
prices of the securities held long. Furthermore, the trading imperatives
of long-short management, which call for keeping dollar
amounts of aggregate longs and aggregate shorts roughly equalized
on an ongoing basis, will tend to limit short-side losses be
shorts are covered as their prices rise; if a gap-up in the price of an
individual security does not afford the opportunity to cover, the
overall portfolio wilstill be protected as long as it is well diversified.
So, the risk represented by the theoretically unbounded losses
on short positions is considerably mitigated in practice.
Myth 12. Long-short portfolios must have more active risk than longonly
portfolios because they take "more extreme" positions.
Because it is not constrained by index weights, a long-short
portfolio may be able to take larger positions in securities with
higher (and lower) expected retums compared with. a long-only
portfolio, which is constrained by index weights. The benefits of
long-short construction, however, do not depend upon the manager's
taking such positions. Integrated optimization will ensure
that long-short selections are made with a view to maximizing expected
return at the risk level at which the client feels most com
able. Given the added flexibility a long-short portfolio affords in
implementation of investment insights, it should be able to improve